The overall objective of the project is to study the epidemiology and pathognomonic development of stuttering in early childhood with special reference to identification of subgroups of stutterers and isolation of factors that are predictive of spontaneous recovery or of chronic stuttering. The specific aims are to test the hypotheses that severity, chronicity and recovery of stuttering are related to (a) symptom pattern, (b)onset type and age at onset, (c) health, (d) articulatory-phonologic and language skills, (e) gender, and (f) underlying genetic liability to the disorder. Such data on the formative stages of stuttering will generate information responding to several important needs. First, it will allow for early diagnosis of stuttering as a clinical entity. Second, it will allow for early identification of high-risk, low-risk, and no-risk incipient stutterers. Third, it will provide a rational basis for decision-making concerning timing of clinical intervention. Fourth, it will promote differentiation in treatment specific to the characteristics of the disorder at each stage. Finally, subgrouping of stuttering will greatly improve all aspects of theoretical and clinical research of stuttering. To accomplish these objectives, 120 young children who stutter will be studied longitudinally from onset of the disorder through a period of four to six years. Many characteristics of stuttering will be observed and recorded periodically. Data will be obtained from medical and developmental records, genetic analyses, analyses of the frequency/type counts of stuttering behaviors in audio and video recordings, severity measures, acoustic analyses of speech signals, tests of phonologic/language skills, anxiety, and nonverbal function. Some comparisons will be made with 40 nonstuttering children as controls. By applying various statistical methods to data obtained from the same subjects at frequent intervals, it will be possible to discern subgroups that emerge along diverging developmental trends in relation to the various parameters under investigation mentioned above. It will also be possible to determine the rate, magnitude, and timing of remission, and make reasonably accurate risk estimates for individual patients (subjects) regarding chances of enduring a chronic disorder.